Thanks Chet, One thing that I'd like to point out as well is that if someone tries this as the code stands then there will be unpredictable results if they try to back out the second data point.
When I tried to take the DERIVE data-points out the graphing of the GAUGE values never resumed. It looked as if the zenperfsnmp was still writing the values to the derive rrd. I had to completely remove the data source and recreate it for it to begin graphing correctly again. -Trey On 12/24/07, Chet Luther <[EMAIL PROTECTED]> wrote: > > I've got something for the group to ponder over the weekend, if you > > wish to that is... I'm trying to track the number of BGP peers on > > our edge routers, and I can get the number and store it as a GAUGE to > > give the appropriate graph as into quantity of peers. However, I'd > > like to also track the rate of change of this number, so I created > > another data point for each data source to be DERIVE. > > > > My problem is this: When I add the second data point, the DERIVE > > graph starts working, however there's no more data plotted for the > > GAUGE graphs. > > > > simplified: > > 1 data source -> 2 data points > > 1 data point -> 1 graph (total 2 graphs for each data source) > > > > I'm only getting graphed output from one data point though. > > This is a known limitation/problem/decision. Up until reading this > post I had never seen a good reason to record the collected value of a > single OID into multiple data points, but this makes complete sense. > I've added ticket #2506 to allow the same value to be recorded in the > RRD file of multiple data sources. > > Ticket #2506: > http://dev.zenoss.org/trac/ticket/2506 > > _______________________________________________ > zenoss-users mailing list > [email protected] > http://lists.zenoss.org/mailman/listinfo/zenoss-users > _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
